help > Is it Safe to Delete Already Processed Subjects
Showing 1-3 of 3 posts
Display:
Results per page:
Jul 12, 2019  12:07 PM | Victoria Okuneye - University of Chicago
Is it Safe to Delete Already Processed Subjects
Hi Alfonso and CONN Users,

I have a few subjects that failed preprocessing due to bad resting state scans.
I left them in to not interfere with order. I continued with ROI maps, Denoising and 1st Level but just didn't select the bad subjects for these steps.
Now when I try to do 2nd level it won't bring up results due to lack or ROI information and such for the bad subjects.

I want to know if it's safe to delete the subjects without redoing the whole preprocessing and analysis stream (which would be cumbersome since this is for hundreds of subjects and ends up using a lot of memory resources from our hcp cluster).

I am worried that deleting the subjects will lead to inaccurate matching of subjects files/covariate information.
Like if I delete subject 500, and subject 501 become 500...... will the files created before the deletion stay with the correct subject going forward and will the 2nd level analysis pull in the right files. I know the conn outputs often say Subject###. 

Insight on this would be greatly appreciated.

Best,
Victoria O
Jul 12, 2019  01:07 PM | Stephen L. - Coma Science Group, GIGA-Consciousness, Hospital & University of Liege
Is it Safe to Delete Already Subjects
> Hello,
>
I can't answer your question, but you can try to set these subjects as nan
in wind level covariates (instead of 0) to completely exclude them from the
results?

Jul 16, 2019  12:07 PM | Victoria Okuneye - University of Chicago
RE: Is it Safe to Delete Already Processed Subjects
Unfortunately, that's something I had already tried and it still won't let me proceed.
Which is why I want to know if it's safe to delete the subjects.